Tracheotomy is a common practice in patients admitted to the Intensive Care Unit (ICU) and subjected to mechanical ventilation.1–3 The technique offers a number of potential advantages over prolonged translaryngeal intubation in terms of patient comfort, a lesser need for sedation, and improved weaning from the ventilator.3,4 One-third of all the tracheotomies are performed in situations where programmed extubation has failed and reintubation is required.3 However, little information is available on the possible benefits of tracheotomy in the initial days after reintubation. The present study analyzes the outcome of patients with tracheotomy after failure of extubation as compared with patients in whom tracheotomy was not performed.

A secondary analysis was performed of an international, multicenter prospective study carried out in 36 ICUs in 8 countries (see Appendix A) between September 2005 and December 2006.5 Given the observational and non-interventional nature of its design, the corresponding Ethics Committee approved the study without the need for informed consent.

From a cohort of 1152 patients subjected to programmed extubation, we analyzed 180 reintubated patients (16%), divided into two groups: 52 patients (29%) tracheotomized after reintubation and 128 patients in principle not subjected to tracheotomy (Fig. 1). The median time from reintubation to tracheotomy was 2.5 days (interquartile range 1.8). Table 1 compares the basal characteristics of the two groups. In the tracheotomy group there were more patients requiring mechanical ventilation due to neurological disease (23% vs 11%, p=0.04), while the non-tracheotomy group was characterized by more patients with pneumonia (19% vs 6%, p=0.03). The patients with tracheotomy had a higher incidence of sepsis and ventilator associated pneumonia after reintubation (Table 1).

Figure 1.

Flow chart of the patients analyzed in the study.

(0.16MB).

The tracheotomy group had a significantly longer stay than the non-tracheotomized patients (Table 2). In the multiple linear regression analysis, three variables were associated with a longer stay in the ICU: days of mechanical ventilation prior to extubation (p<0.001), post-reintubation sepsis (p<0.001) and tracheotomy (p=0.025).

The mortality rate in the ICU proved similar in both groups: 31% in the tracheotomy group versus 27% in the non-tracheotomy group (relative risk of tracheotomy: 1.16; 95% confidence interval: 0.70–1.90; p=0.57).

The mortality rate was similar in the patients with early tracheotomy (in the first 3 days after reintubation) and in those in which tracheotomy was performed after the third post-extubation day: 32% versus 29% (p=0.82).

The main finding of this observational study is that tracheotomy in patients who have required reintubation does not offer advantages in terms of the duration of stay in the ICU or mortality as compared with patients in which tracheotomy is not performed after the first reintubation.

In recent years there has been a significant increase in the number of tracheotomies performed in patients subjected to mechanical ventilation,6 this being partially conditioned by introduction of the percutaneous technique in routine clinical practice.7,8 Tracheotomy, in comparison to prolonged translaryngeal intubation, has demonstrated advantages in terms of patient comfort, a lesser need for sedation, oral intake, improved oral hygiene, or a faster recovery of speech.9 There is greater controversy regarding its possible advantages related to other clinical outcomes such as the incidence of ventilator associated pneumonia,10 faster weaning from mechanical ventilation,11 patient safety12,13 or, most importantly, mortality and the duration of stay in the ICU.

A recent metaanalysis14 including 7 randomized clinical trials comparing early tracheotomy versus late tracheotomy or prolonged intubation has found that early tracheotomy does not reduce mortality during the first 90 days of stay (relative risk: 0.86; 95% confidence interval: 0.65–1.13). Early tracheotomy likewise is not associated with a shortening of stay in the ICU (weighted mean difference: −6.93 days; 95% confidence interval: −16.50 to 2.63).

In addition to either early or late, tracheotomy can be classified in relation to the course of mechanical ventilation: during ventilatory support, during weaning from mechanical ventilation or after the failure of extubation and the start of a new episode of mechanical ventilation. In a previous analysis by our group, we found that in one-third of the patients tracheotomy was performed after the failure of extubation.5 To date, no study had evaluated the effect of tracheotomy in the subgroup of patients who are tracheotomized after reintubation due to post-extubation respiratory failure.

Our study has shown that independently of other variables, patients with tracheotomy after reintubation had a longer stay. In contrast to other observational studies4,15–18 in which tracheotomized patients were reported to have lesser mortality in the ICU, in our patient cohort the mortality rate was similar in both groups–this observation probably being related to differences in the populations included in the studies.

Our study has a number of limitations. Firstly, the limited sample size may have accounted for some of the observations, and precluded the use of more sophisticated statistical analyses such as for example a trend index to determine the individual probability of tracheotomy, which would have allowed a better estimation of the effect of tracheotomy upon the primary outcome variable. Secondly, the observational, non-interventional nature of the study implies that neither the decision to reintubate nor the decision to perform tracheotomy was protocolized, and was left to the criterion of the physician in charge of the patient. In this context, there may have been selection bias, since on one hand tracheotomy was more frequent in patients in which coma constituted the indication of ventilation–this population presenting a poorer prognosis19–while on the other hand tracheotomy is usually decided in patients considered to have a more favorable vital prognosis. In our study there were no differences in severity upon admission to the ICU, though no severity index was recorded at the time of reintubation or at the time of tracheotomy.

In conclusion, we have not found tracheotomy after failed extubation to offer advantages over the maintenance of translaryngeal intubation. In any case, these findings must be considered to serve the purpose of generating hypotheses, and the true effect of tracheotomy after reintubation must be evaluated in the context of a randomized, prospective and controlled clinical trial.

This study has been funded by the CIBER de Enfermedades Respiratorias, Instituto Carlos III, Spain.

The authors declare that they have no conflicts of interest.